Skip to main content
Method · Reading paths

Four curated sequences

The archive read in the order that makes each argument. Every path is a short reading sequence — four or five pieces — that prepares you to pick up one of the four playbooks.

PathReviewed Apr 2026 · 7 months ago

Introducing AI-Assisted Dev to a New Team

A curated five-post sequence that takes you from "we should be using AI more" to a team practice that survives contact with production.

For: Engineering leaders and senior ICs joining a team or rolling out AI assistants for the first time.~17 min total
  1. Start with the first-quarter rollout: agree on data and permission rules, define a pilot, establish checks, and review the results.

  2. Use the one-page policy to agree on ownership, data access, verification, and escalation before the pilot.

  3. Keep time for independent diagnosis and review, especially when an assistant handles unfamiliar code.

Prepares you for the playbook
AI-Assisted Dev Adoption Loop →

How a team moves from "we tried Copilot" to a reproducible AI-assisted development practice without losing engineering judgment.

PathReviewed Apr 2026 · 7 months ago

Auditing and Stabilizing GPU Infrastructure

A five-post sequence for platform leads who need to turn GPU assumptions into baselines and ship a credible 90-day roadmap.

For: Platform and infrastructure teams operating GPU workloads or planning to take them on.~0 min total
    Prepares you for the playbook
    AI Infrastructure Readiness Audit →

    A repeatable process for auditing production AI infrastructure: what to measure, what to fix first, and how to write down the answer.

    PathReviewed Apr 2026 · 7 months ago

    Healthcare Integration From Scratch

    A four-part sequence that takes you from national interoperability context to a specific failure mode to the reference implementation.

    For: Integration engineers, clinical data teams, and platform leads responsible for partner feeds.~27 min total
    1. Start with the national context. What changed between Meaningful Use and TEFCA, and what still blocks plug-and-play apps despite a decade of standards work.

    2. The operational playbook: contracts as assets, observability first-class, executable docs, and support workflows that compound.

    3. A specific failure mode — an "enhanced best match" trap where filter and mode parameters collide — and the defensive pattern that fixes it on both sides.

    4. The reference implementation. How Source Profiles, a three-phase parsing pipeline, and a workflow DSL turn messy legacy formats into semantic events without a one-off codebase per feed.

    Prepares you for the playbook
    Healthcare Integration Onboarding →

    How to add a new healthcare feed — HL7v2, CSV, EDI, FHIR, or CDA — without a one-off codebase and without silent data loss.

    PathReviewed Apr 2026 · 7 months ago

    Rolling Out Production Agents

    A four-post sequence for teams moving agents from "impressive demo" to "owns a class of work with visible track record."

    For: Teams piloting or scaling AI agents inside an engineering organization.~9 min total
    1. Maintain engineering judgment through independent checks, explanation, and occasional work without the assistant.

    2. The policy that makes the practice auditable. Ten lines the team converges on, with a quarterly review rhythm.

    Prepares you for the playbook
    Agent Rollout With Guardrails →

    A staged path from "one agent in one repo" to "agents own a class of work" without inheriting opaque state or runaway tool calls.

    Different audience or question?

    The paths above are the four I have written down so far. If your team needs a different sequence — for example, a reliability-only path or a compliance-oriented one — I am happy to sketch it.

    Reading Paths | Cody Blevins